Security and fraud prevention techniques based on detection of a transaction card in an image

Patent metadata

Field	Value
Publication number	US-10872344-B2
Application number	US-201916283395-A
Country	US
Kind code	B2
Filing date	Feb 22, 2019
Priority date	Feb 22, 2019
Publication date	Dec 22, 2020
Grant date	Dec 22, 2020

Abstract

In some implementations, a transaction card security system may detect a transaction card in an image, and may perform image processing to identify one or more character sequences or one or more designs that appear on the transaction card. The one or more character sequences or the one or more designs may identify an institution associated with the transaction card and/or a cardholder associated with the transaction card. The transaction card security system may identify the institution or the cardholder associated with the transaction card based on the one or more character sequences or the one or more designs, and may perform one or more actions relating to security of the transaction card based on detecting the transaction card in the image and identifying the institution or the cardholder associated with the transaction card.

First claim

What is claimed is:

1. A method, comprising:
performing, by a system, image processing on a set of images in an image repository,
wherein the set of images are input into a machine learning model, executing on the system, that has been trained to recognize transaction cards in images, and
wherein the machine learning model is trained based on at least one of:
a training image and a mask image,
wherein the mask image identifies a first region that includes a particular transaction card and a second region that does not include the particular transaction card, or

information that identifies pixels, in the training image, that include the particular transaction card;

detecting, by the system and based on performing the image processing, a transaction card in an image of the set of images;
determining, by the system and based on analyzing the image, at least one of:
a design on the transaction card,
a name on the transaction card, or
a card number on the transaction card;

identifying, by the system, at least one of an organization or a cardholder associated with the transaction card based on at least one of:
the design on the transaction card,
the name on the transaction card, or
the card number on the transaction card; and

performing, by the system, an action relating to security of the transaction card based on detecting the transaction card in the image and identifying at least one of the organization or the cardholder,
the action including one or more of:
blurring a portion of the image identifying the transaction card,
redacting the portion of the image identifying the transaction card, or
replacing the portion of the image identifying the transaction card.

2. The method of claim 1 , wherein the set of images are associated with a social media account.

3. The method of claim 1 , wherein identifying the organization comprises:
identifying a financial institution associated with the card number.

4. The method of claim 1 , wherein performing the action comprises:
sending a message to at least one of the organization or the cardholder,
the message including at least one of:
the name,
a portion of the name,
the card number, or
a portion of the card number.

5. The method of claim 1 , wherein performing the action comprises at least one of:
placing the transaction card on hold,
cancelling the transaction card,
generating a new card number to replace the card number of the transaction card,
issuing a new transaction card, to the cardholder, to replace the transaction card,
sending an instruction to delete the image from a social media account associated with the image repository,
sending a message, to the social media account, regarding the image, or
sending a message, to a social media service provider associated with the social media account, regarding the image.

6. A device, comprising:
a memory; and
one or more processors, communicatively coupled to the memory, configured to:
perform image processing on a set of images in an image repository,
wherein the set of images are input into a machine learning model, executing on the device, that has been trained to recognize transaction cards in images, and
wherein the machine learning model is trained based on at least one of:
a training image and a mask image,
 wherein the mask image identifies a first region that includes a particular transaction card and a second region that does not include the particular transaction card, or
 information that identifies pixels, in the training image, that include the particular transaction card;

detect, based on performing the image processing, a transaction card in an image of the set of images;
process the image to identify one or more character sequences or one or more designs on the transaction card;
determine an entity associated with the transaction card based on at least one of:
a character sequence, of the one or more character sequences, or
a design of the one or more designs;

perform a first action based on the entity,
the first action relating to security of the transaction card; and

perform a second action based on detecting the transaction card in the image,
the second action including at least one of:
blurring a portion of the image identifying the transaction card,
redacting the portion of the image identifying the transaction card, or
replacing the portion of the image identifying the transaction card.

7. The device of claim 6 , wherein the image is obtained from social media.

8. The device of claim 6 , where the one or more processors, when performing the first action based on the entity, are configured to:
perform a first particular action when the entity is a first financial institution, or
perform a second particular action when the entity is a second financial institution,
wherein the second particular action is different from the first particular action when the second financial institution is different from the first financial institution.

9. The device of claim 8 , wherein the first particular action includes sending a message to the first financial institution, and
wherein the second particular action includes sending a message to a cardholder associated with the second financial institution,
wherein the cardholder is identified based on the one or more character sequences.

10. The device of claim 6 , wherein the one or more processors are further configured to:
perform a third action including at least one of:
placing the transaction card on hold,
cancelling the transaction card,
flagging one or more transactions made using the transaction card,
generating a new card number to replace a card number of the transaction card, or
issuing a new transaction card to replace the transaction card.

11. The device of claim 6 , wherein the one or more processors are further configured to:
determine a location associated with the transaction card; and
perform a third action based on the location.

12. The device of claim 11 , wherein the location is determined based on at least one of:
recognition of an object included in the image,
recognition of text included in the image,
information included in a social media post that includes the image, or
a time that the image was posted to social media and a proximate time of a transaction for which the transaction card was used.

13. The device of claim 11 , wherein the one or more processors, when performing the third action, are configured to at least one of:
indicate the location in a message to be sent to the entity, or
update a risk score associated with the location.

14. The device of claim 6 , wherein the one or more processors are further configured to:
update a risk score associated with a cardholder of the transaction card.

15. A non-transitory computer-readable medium storing instructions, the instructions comprising:
one or more instructions that, when executed by one or more processors, cause the one or more processors of a system to:
perform image processing on a set of images in an image repository,
wherein the set of images are input into a machine learning model, executing on the system, that has been trained to recognize transaction cards in images, and
wherein the machine learning model is trained based on at least one of:
a training image and a mask image,
 wherein the mask image identifies a first region that includes a particular transaction card and a second region that does not include the particular transaction card, or
 information that identifies pixels, in the training image, that include the particular

Assignees

• Capital One Services Llc

Inventors

• Farivar Reza
• Kubiske Michael
• Goodsitt Jeremy
• Watson Mark

Classifications

• G06Q20/409 Primary

  Device specific authentication in transaction processing · CPC title

• G06V20/63

  Scene text, e.g. street names · CPC title

• G06Q20/4016 Primary

  involving fraud or risk level assessment in transaction processing · CPC title

• G06V30/10

  Character recognition · CPC title

• G06V20/62

  Text, e.g. of license plates, overlay texts or captions on TV images · CPC title